CN112001738A - Method for constructing multi-factor logistics product price model and application method thereof - Google Patents
Method for constructing multi-factor logistics product price model and application method thereof Download PDFInfo
- Publication number
- CN112001738A CN112001738A CN201910446921.3A CN201910446921A CN112001738A CN 112001738 A CN112001738 A CN 112001738A CN 201910446921 A CN201910446921 A CN 201910446921A CN 112001738 A CN112001738 A CN 112001738A
- Authority
- CN
- China
- Prior art keywords
- data
- factor
- city
- price
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0206—Price or cost determination based on market factors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a method for constructing a multi-factor logistics product price model and an application method thereof, wherein the construction method comprises the following steps: acquiring city related factor data and first price data and first quantity data of at least one logistics product in a flow direction; the city correlation factor data comprises economic indexes of at least two cities, the distance between any two cities and city grades; training and learning by utilizing the city correlation factor data to obtain a multi-factor coefficient model so as to output multi-factor coefficients; and training and learning by utilizing the first price data, the first quantity data and the multi-factor coefficient to obtain a logistics product price model. According to the method for constructing the multi-factor logistics product price model, the logistics product price model is constructed by obtaining the plurality of city related factor data, and the prediction value of the constructed logistics product price model can be improved.
Description
Technical Field
The invention relates to the technical field of logistics, in particular to a method for constructing a multi-factor logistics product price model and an application method thereof.
Background
With the rapid development of the logistics industry, the future competitiveness of enterprises is directly influenced by formulating a set of price scheme meeting the market. The price model is used as a core part of a pricing system and directly influences the formulation of a final price scheme.
At present, a demand model in international economics is mainly adopted for a price model, and differences among cities enable the prediction value of the existing price model to be not high when the existing price model is used for evaluating the price relations of different cities.
Disclosure of Invention
In view of the above-mentioned drawbacks and deficiencies of the prior art, it is desirable to provide a method for constructing a multi-factor logistics product price model and a method for applying the same.
In a first aspect, the present invention provides a method for constructing a multi-factor logistics product price model, including:
acquiring city related factor data and first price data and first quantity data of at least one logistics product in a flow direction; the city correlation factor data comprises economic indexes of at least two cities, the distance between any two cities and city grades;
training and learning by utilizing the city correlation factor data to obtain a multi-factor coefficient model so as to output multi-factor coefficients;
and training and learning by utilizing the first price data, the first quantity data and the multi-factor coefficient to obtain a logistics product price model.
In one embodiment, training and learning to obtain a multi-factor coefficient model by using city correlation factor data includes:
preprocessing city correlation factor data;
and training and learning by utilizing the preprocessed city correlation factor data to obtain a multi-factor coefficient model.
In one embodiment, preprocessing the city relevance factor data includes at least one of:
filling null values into the city related factor data;
abnormal value detection is carried out on the data of the city correlation factors;
and carrying out scaling processing on the city correlation factor data.
In one embodiment, the scaling process for the city correlation factor data includes:
carrying out normalization processing on the city related factor data;
and carrying out scaling processing on the normalized city correlation factor data so that the scaled structure and the first price data belong to the same range.
In one embodiment, training and learning the preprocessed city correlation factor data to obtain a multi-factor coefficient model, including:
determining whether a competition impact factor is present;
if not, training and learning by utilizing the preprocessed city related factor data to obtain a first multi-factor coefficient model so as to output multi-factor coefficients;
if yes, obtaining second price data corresponding to the competition influence factors;
and training and learning by utilizing the preprocessed city correlation factor data and the second price data to obtain a second multi-factor coefficient model so as to output multi-factor coefficients.
In one embodiment, before training and learning the logistics product price model by using the first price data, the quantity data and the multi-factor coefficient, the method further comprises the following steps:
and carrying out logarithmic transformation on the first price data and the first volume data.
In a second aspect, the invention provides a method for applying a multi-factor logistics product price model, which comprises the following steps:
receiving first price data or a first piece of volume data;
and determining the component data corresponding to the first price data or the price data corresponding to the first component data according to a functional mapping relation between the price data and the component data, wherein the functional mapping relation is that the product of the price elasticity coefficient power of the price data and the multi-factor coefficient is used for representing the component data.
In one embodiment, the multi-factor coefficients are determined by a pre-constructed multi-factor coefficient model.
In one embodiment, the multi-factor coefficient model is constructed by:
acquiring city correlation factor data, wherein the city correlation factor data comprises economic indexes of at least two cities, a distance between any two cities and city grades;
and training and learning by using the city correlation factor data to obtain a multi-factor coefficient model.
In one embodiment, training and learning to obtain a multi-factor coefficient model by using city correlation factor data includes:
determining whether a competition impact factor is present;
if not, training and learning by utilizing the city related factor data to obtain a first multi-factor coefficient model so as to output a multi-factor coefficient;
if yes, obtaining second price data corresponding to the competition influence factors;
and training and learning by utilizing the city correlation factor data and the second price data to obtain a second multi-factor coefficient model so as to output a multi-factor coefficient.
According to the method for constructing the multi-factor logistics product price model and the application method thereof, the multi-factor coefficient model is obtained through training and learning by utilizing the acquired multiple city related factor data to output the multi-factor coefficient, and then the logistics product price model is obtained through training and learning by utilizing the acquired first price data, price data and the multi-factor coefficient. The method for constructing the multi-factor logistics product price model provided by the embodiment obtains the multiple city related factor data, the first price data and the piece data in each flow direction, can highlight the difference between the first price data and the piece data in each flow direction, and adjusts the first price data and the piece data in different flow directions through the multiple city related factor data, so that the prediction value of the obtained logistics product price model is improved.
Further, according to the embodiment of the application, the prediction value of the obtained logistics product component model is higher by preprocessing the city related factor data.
Further, the multi-factor coefficient corresponding to whether the competitive influence factor exists or not is obtained in the embodiment, so that the prediction value of the logistics product price model obtained by utilizing the multi-factor coefficient training learning is higher.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:
FIG. 1 is a schematic flow chart of a method for constructing a multi-factor logistics product price model according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of step 120 provided by the embodiment of the present invention;
FIG. 3 is a schematic flow chart of a method for applying a multi-factor logistics product price model according to an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of an apparatus for constructing a multi-factor logistics product price model according to an embodiment of the present invention;
FIG. 5 is a schematic structural diagram of an apparatus for constructing a multi-factor logistics product price model according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of a computer device according to an embodiment of the present invention.
Detailed Description
The present application will be described in further detail with reference to the following drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant invention and not restrictive of the invention. It should be noted that, for convenience of description, only the portions related to the present invention are shown in the drawings.
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.
The traditional price model mainly adopts a demand model in international economics, but due to differences among cities, the predicted price of the price model is not high.
Therefore, it is desirable to provide a method for constructing a multi-factor logistics product price model, which is used for improving the prediction value of the price model.
According to the method, the influence factors are screened according to experience, the relationship among the factors is constructed, a corresponding learning model is formed, actual market application is carried out, practice verifies are carried out, prediction is accurate and convenient, the existing pure artificial experience decision is not relied on, and errors and subjectivity different from person to person are avoided.
Referring to fig. 1, an exemplary flow chart of a method of constructing a multi-factor logistics product price model is shown as described in one embodiment of the present application.
As shown in fig. 1, in step 110, city correlation factor data and first price data and first quantity data of at least one flowing logistics product are obtained; the city correlation factor data comprises economic indexes of at least two cities, distances between any two cities and city grades.
A flow direction refers to a parcel flowing from an originating city to a destination city, for example, a parcel is in a flow direction from Beijing to Shenzhen, and a parcel is in another flow direction from Guangzhou to Shenzhen.
The logistics product refers to express delivery products adopting different transportation modes, for example, the logistics product is transported by air, and the logistics product is transported by land. Since the difference between the first price data and the first quantity data obtained by day for each flow direction of the logistics products is not great, a relatively long time span, such as the month, can be considered to obtain the first price data and the first quantity data for each flow direction of the logistics products. Wherein the first price data refers to the average price of the logistics products in each flow direction every month.
The city correlation factor data refers to information data related to cities, and optionally, the city correlation factor data may include economic indexes of at least two cities, a distance between any two cities, a city grade, and the like. The reason for acquiring the city correlation factor data is that the price of the city is different due to the fact that the economic level difference among different cities is considered, and in order to enable the constructed price model to have higher prediction value, the difference among different cities is added into the price model. It should be noted that the city correlation factor data may be acquired through a network, may also be acquired through other devices, and may also be acquired through other manners, which is not limited herein.
In step 120, the city correlation factor data is used to train and learn the multi-factor coefficient model to output the multi-factor coefficient.
The multi-factor coefficient model is a model for determining multi-factor coefficients. For example, the acquired data such as the city correlation factor data is input into the multi-factor coefficient model, and then the corresponding multi-factor coefficient can be obtained. Wherein the multi-factor coefficient is used for controlling the influence of price change caused by city difference on the quantity of the part. The method comprises the following steps of training and learning to obtain a logistics product price model by utilizing multi-factor coefficients, first price data and first quantity data, and the constructed logistics product price model can be more accurate and the prediction value of the logistics product price model is better improved through the multi-factor coefficients.
When the multi-factor coefficient model is obtained through training and learning, data such as original city related factor data and the like can be directly used for training and learning, or data such as preprocessed city related factor data and the like can be used for training and learning, and the embodiment shown in fig. 2 is that the multi-factor coefficient model is obtained through training and learning by using the preprocessed city related factor data and the like. Further, fig. 2 shows a schematic flowchart of step 120 provided in an embodiment of the present application.
As shown in fig. 2, the step 120 of training and learning to obtain the multi-factor coefficient model by using the city correlation factor data includes:
and step 210, preprocessing the city correlation factor data.
Wherein preprocessing the city correlation factor data comprises at least one of:
filling null values into the city related factor data;
abnormal value detection is carried out on the data of the city correlation factors;
and carrying out scaling processing on the city correlation factor data.
Since the city correlation factor of each city is not obtained, null filling is required to be performed on the city correlation factor data. For example, when city-related factor data of a certain city or some cities is missing (here, mainly, the economic index and the city grade of the city are missing), the city-related factor data of the same city grade of the same province is used for filling, if the city grade is missing, the city economic index of the adjacent city of the same province or the city around the region is used for filling, if the city-related factor data of the whole province is completely missing, the city-related factor data of the province which is the same grade or close to the missing province is searched nationwide, and the searched city-related factor data is used for filling. The present application is not limited to the filling method.
The abnormal value detection of the city related factor data can adopt conventional abnormal detection methods such as an isolated forest abnormal detection method and the like, and the adopted abnormal detection methods are all existing mature technologies and are not described herein again. Abnormal value detection is carried out on the city related factor data, and the abnormal city related factor data are detected, so that the accuracy of the multi-factor coefficient model is improved, and the prediction value of the logistics product price model is improved.
In one embodiment, the scaling process is performed on the city correlation factor data, and comprises the following steps: carrying out normalization processing on the city related factor data; and carrying out scaling processing on the normalized city correlation factor data so that the scaling processing result and the first price data belong to the same range.
The method for normalizing the maximum value and the minimum value is adopted to normalize the city related factor data, and specifically comprises the following steps:
wherein, f (i) is the data of each city correlation factor to be normalized, wherein, i is 1The number f' (i) is the normalized data of each city correlation factor f (i)maxIs the maximum value of the corresponding city correlation factor data, f (i)minIs the minimum value of the corresponding city correlation factor data.
Scaling the normalized city correlation factor data to make the scaling result and the first price data belong to the same range, specifically:
wherein, f (i) _ scale is the relevant factor data of each city after scaling treatment, pmaxIs the maximum value in the first price data of all flow directions, pminF' (i) is the normalized data of each city correlation factor, which is the minimum value in the first price data of all the flow directions.
In this embodiment, the city related factor data is normalized, so that each city related factor data is scaled to the same scale range, for example, the data can be scaled to 0-1 range, the normalized city related factor data is concentrated in a relatively concentrated manner and is concentrated in a range of 0-1, that is, the normalized city related factor data is relatively stable, the multi-factor model obtained by the subsequent steps through training and learning according to the normalized city related factor data is more accurate, and the logistics product price model obtained through training and learning by using the multi-factor coefficient is more predictive.
Further, in this embodiment, the normalized city correlation factor data is scaled to the same range as the first price data, so that the fitting effect between the first volume data can be ensured, and the robustness of the constructed logistics product price model can be ensured.
And step 220, training and learning by utilizing the preprocessed city correlation factor data to obtain a multi-factor coefficient model.
The multi-factor coefficient model is a model for determining multi-factor coefficients.
In the embodiment, the city related factor data is preprocessed, and then the preprocessed city related factor data is used for training and learning to obtain the multi-factor coefficient model, so that the multi-factor coefficient obtained by the multi-factor coefficient model is more accurate, and the prediction value of the logistics product quantity model obtained by training and learning of the multi-factor coefficient is higher.
In step 130, a logistics product price model is obtained by training and learning by using the first price data, the first quantity data and the multi-factor coefficient.
Specifically, preprocessing such as filling and abnormal value monitoring is performed on the acquired first price data and first volume data.
The first price data and the first volume data are filled, for example, when the first price data or the first volume data of a certain flow direction or a certain flow direction is missing, the first price data or the first volume data of the flow direction with the same economic index (the same economic index includes the economic index within the deviation range of the economic index) and the same distance between two cities are filled. When there is a competitor in the logistics product, the first price data or the first volume data can be filled, and second price data with the same level of flow direction as that of the competitor (if there are a plurality of competitors, the second price data of the competitor can be an average value of a plurality of competitor price data with the same level of flow direction) or volume can be used for filling. The present application is not limited to the filling method.
When the logistics product has an absolute dominance in the market, that is, the logistics product only has a certain company or enterprise in the market, or even other companies have the logistics product in the market, but the price of the logistics product is completely dominated by the logistics product of a certain company or enterprise, it means that no competitor exists in the logistics product, otherwise, the logistics product exists in the competitor.
The abnormal value detection is performed on the first price data or the first volume data, and conventional abnormal detection methods such as an isolated forest abnormal detection method and the like can be adopted, wherein the adopted abnormal detection methods are all existing mature technologies and are not described herein again. Abnormal value detection is carried out on the first price data or the first volume data, and the abnormal first price data or the abnormal first volume data are detected, so that the prediction value of the logistics product price model is improved.
Analyzing the histograms of the preprocessed first price data and the preprocessed first quantity data to find that the histogram distribution of the directly acquired original first price data and the original first quantity data is not in accordance with the actual situation, so that before a logistics product price model is obtained by training and learning by using the first price data, the first quantity data and a multi-factor coefficient, logarithmic transformation is performed on the first price data and the first quantity data, and the method specifically comprises the following steps:
pl＝log(p+1)
vl＝log(v+1)
wherein p is the original first price data, plFor the transformed first price data, v is the original first volume data, vlIs the first piece of transformed data.
And carrying out logarithmic transformation on the first price data and the first quantity data, and training and learning the logistics product price model obtained by utilizing the logarithmically transformed first price data and the first quantity data in the subsequent steps to have good robustness and high prediction value.
In this embodiment, according to the obtained multiple city correlation factor data, training and learning are performed to obtain a multiple factor coefficient model to output multiple factor coefficients, the multiple factor coefficients are combined with the obtained first price data and first quantity data of the logistics product in each flow direction, and training and learning are performed to obtain a logistics product price model.
Further, training and learning by utilizing the first price data, the first quantity data and the multi-factor coefficient to obtain a logistics product price model, which specifically comprises the following steps: and training and learning by using the transformed first price data, the transformed first quantity data and the multi-factor coefficient to obtain a logistics product price model.
The method comprises the steps of obtaining historical first price data and historical first volume data, preprocessing and carrying out logarithmic transformation on the obtained historical first price data and historical component data, and obtaining transformed first price data and transformed first volume data. It should be noted that, the historical data in recent years, for example, the last two years, is acquired preferentially, otherwise, the acquired data has no actual prediction value in the future.
The transformed first price data and the transformed first quantity data are divided into training data and test data, and the ratio of the training data to the test data is 7: 3.
The logistics product price model constructed according to the transformed first price data, the transformed first quantity data and the multi-factor coefficient is formula (1):
vl＝k*pl e1 (1)
wherein p islFor the transformed first price data, vlFor the transformed first piece of volume data, k is the multi-factor coefficient and e1 is the price elasticity coefficient.
The multi-factor coefficient is determined according to a multi-factor coefficient model obtained by training and learning by utilizing the preprocessed city related factor data.
In one embodiment, training and learning the preprocessed city correlation factor data to obtain a multi-factor coefficient model, including:
determining whether a competition impact factor is present;
if not, training and learning by utilizing the preprocessed city related factor data to obtain a first multi-factor coefficient model so as to output multi-factor coefficients;
if yes, obtaining second price data corresponding to the competition influence factors;
and training and learning by utilizing the preprocessed city correlation factor data and the second price data to obtain a second multi-factor coefficient model so as to output multi-factor coefficients.
Wherein, the competition influence factor refers to competitors who comprise the same logistics products.
If no competitor exists, training and learning by utilizing the preprocessed city related factor data to obtain a first multi-factor coefficient model so as to output a multi-factor coefficient, which is specifically a formula (2):
wherein k is1F (i) _ scale is preprocessed city correlation factor data, wherein i is 1iIs the weight of each city correlation factor data.
Acquiring relevant factor data, first price data and first quantity data of each historical city, dividing the acquired data into training data and testing data respectively, and setting the ratio of the training data to the testing data to be 7: 3. Multiplying factor k in formula (2)1Substituting the expression(s) into the formula (1), and fitting the training data based on the formula obtained by combining the formula (1) and the formula (2) and by using the obtained city correlation factor data, the first price data and the first quantity data by using a quasi-Newton method to obtain the weight beta of each corresponding city correlation factor data when no competitor existsiAnd a price elastic coefficient e 1. Weighting beta of each city correlation factor dataiAnd substituting the formula (2) into a first multi-factor coefficient model, and substituting the first multi-factor coefficient model and the price elastic coefficient e1 into the formula (1) into a logistics product price model without a competitor of the logistics product. It should be noted that the method for obtaining the parameters by fitting is not limited here.
And testing the obtained logistics product price model by using the obtained test data of the city correlation factor data, the first price data and the first quantity data, checking various indexes such as accuracy, recall rate, AUC and the like to evaluate the logistics product price model, if the indexes are in a threshold range, using the obtained logistics product price model, and otherwise, obtaining historical data again to construct the logistics product price model until the indexes are in the threshold range.
The second price data is price data corresponding to a competitor, and further, if there are multiple competitors, the second price data may be an average value of the price data of the multiple competitors.
If a competitor exists, training and learning by utilizing the preprocessed city correlation factor data and the second price data to obtain a second multi-factor coefficient model so as to output multi-factor coefficients, wherein the method specifically comprises the following steps:
wherein k is2F (i) _ scale is preprocessed city correlation factor data, wherein i is 1iFor the weight of each city correlation factor data, p' is the second price data, and e2 is the weight of the competitor.
Obtaining relevant factor data of each historical city, first price data, first quantity data and second price data, dividing the obtained data into training data and testing data respectively, and setting the proportion of the training data to the testing data to be 7: 3. Multiplying factor k in formula (3)2Substituting the expression(s) into the formula (1), and fitting by using the obtained training data of the city correlation factor data, the first price data, the first volume data and the second price data based on the formula obtained by combining the formula (1) and the formula (3) to obtain the weight beta of each city correlation factor data corresponding to the existence of a competitor by adopting a conjugate gradient methodiPrice elastic coefficient e1 and competitor's weight e 2. Weighting beta of each city correlation factor dataiSubstituting the weight e2 of the competitor into the formula (3) to obtain a second multi-factor coefficient model, and substituting the second multi-factor coefficient model and the price elasticity coefficient e1 into the formula (1) to obtain the competitor with logistics productsThe logistics product price model of (1). It should be noted that the method for obtaining the parameters by fitting is not limited here.
It should be noted that the obtained second price data may be preprocessed, including filling, outlier detection, and scaling, and the preprocessing method is the same as the preprocessing method for the first price data, and is not described here again.
And testing the obtained logistics product price model by using the obtained test data of the city correlation factor data, the first price data, the first quantity data and the second price data, checking various indexes such as accuracy, recall rate, AUC and the like to evaluate the logistics product price model, if the indexes are in a threshold range, using the obtained logistics product price model, and otherwise, obtaining historical data again to construct the logistics product price model until the indexes are in the threshold range.
In this embodiment, the multi-factor coefficient model is obtained by training and learning under two conditions of whether there are competing influence factors, so as to obtain the multi-factor coefficients corresponding to the two conditions. The multi-factor coefficients corresponding to the two conditions are obtained in the embodiment, so that the prediction value of the logistics product price model obtained by the subsequent step of training and learning by using the multi-factor coefficients is higher.
Referring to fig. 3, an exemplary flow chart of a multi-factor logistics product price model application method is shown as described in accordance with one embodiment of the present application.
As shown in FIG. 3, in step 310, a first price data or a first piece of volume data is received.
Specifically, the multi-factor logistics product price quantity model receives first price data or first quantity data. Wherein the first price data refers to the average price of the logistics products in each flow direction every month. The first volume data refers to the volume of the respective flow direction of the logistics product per month.
As shown in fig. 3, in step 320, the component data corresponding to the first price data or the price data corresponding to the first component data is determined according to a functional mapping relationship between the price data and the component data, wherein the functional mapping relationship is a product of a price elasticity coefficient power of the price data and a multi-factor coefficient used for characterizing the component data.
Specifically, the function mapping relationship may be expressed as:
vl＝k*pl e1
wherein p islAs price data, vlFor the piece data, k is the multi-factor coefficient and e1 is the price elasticity coefficient. Wherein, the price data and the component data can be the price data and the component data after being converted by logarithm. The multi-factor coefficient may be preset, or may be obtained through the following embodiments, which are not limited herein.
When the multi-factor logistics product price model receives the first price data in step 310, step 320 obtains the quantity data corresponding to the first price data according to the function mapping relation between the price data and the quantity data. When the multi-factor logistics product price quantity model receives the first quantity data in step 310, the price data corresponding to the first quantity data is obtained in step 320 according to the functional mapping relation between the price data and the quantity data.
In this embodiment, according to the functional mapping relationship between the price data and the component data, the price data or the component data can be predicted more accurately.
In one embodiment, the multi-factor coefficients may be determined by a pre-constructed multi-factor coefficient model. The pre-constructed multi-factor coefficient model may be already constructed on other devices, or may be constructed through the following embodiments. This is not a limitation here.
In one embodiment, the multi-factor coefficient model is constructed by:
acquiring city correlation factor data, wherein the city correlation factor data comprises economic indexes of at least two cities, a distance between any two cities and city grades;
and training and learning by using the city correlation factor data to obtain a multi-factor coefficient model.
Specifically, the city correlation factor data refers to information data related to cities, and optionally, the city correlation factor data may include economic indexes of at least two cities, a distance between any two cities, a city grade, and the like. The reason for acquiring the city correlation factor data is that the price of the city is different due to the fact that the economic level difference among different cities is considered, and in order to enable the constructed price model to have higher prediction value, the difference among different cities is added into the price model. It should be noted that the city correlation factor data may be acquired through a network, may also be acquired through other devices, and may also be acquired through other manners, which is not limited herein.
When the multi-factor coefficient model is obtained by training and learning the city related factor data, preprocessing including null filling, abnormal value detection, scaling processing and the like can be firstly carried out on the city related factor data. The specific preprocessing process has already been stated above and is not described in detail here.
In one embodiment, training and learning to obtain a multi-factor coefficient model by using the city correlation factor data comprises:
determining whether a competition impact factor is present;
if not, training and learning by utilizing the city related factor data to obtain a first multi-factor coefficient model so as to output a multi-factor coefficient;
if yes, obtaining second price data corresponding to the competition influence factors;
and training and learning by utilizing the city correlation factor data and the second price data to obtain a second multi-factor coefficient model so as to output a multi-factor coefficient.
Specifically, how to train and learn to obtain the multi-factor coefficient model by using the city correlation factor data to output the multi-factor coefficient under the two conditions of whether there are competing influence factors or not is specifically described in the foregoing embodiments, and is not described here again.
Fig. 4 is a schematic structural diagram of an apparatus 400 for constructing a multi-factor logistics product price model according to an embodiment of the present invention. As shown in fig. 4, the apparatus may implement the method shown in fig. 1, and the apparatus may include:
the acquisition module 410 is used for acquiring city related factor data and first price data and piece data of at least one logistics product in a flow direction; the city correlation factor data comprises economic indexes of at least two cities, the distance between any two cities and city grades;
the first building module 420 is configured to train and learn to obtain a multi-factor coefficient model by using the city correlation factor data to output a multi-factor coefficient;
and the second building module 430 is used for training and learning to obtain a logistics product price model by using the first price data, the quantity data and the multi-factor coefficient.
Optionally, the first building module 420 is further configured to:
preprocessing city correlation factor data;
and training and learning by utilizing the preprocessed city correlation factor data to obtain a multi-factor coefficient model.
Optionally, preprocessing the city-related factor data comprises at least one of:
filling null values into the city related factor data;
abnormal value detection is carried out on the data of the city correlation factors;
and carrying out scaling processing on the city correlation factor data.
Optionally, the scaling process is performed on the city correlation factor data, and includes:
carrying out normalization processing on the city related factor data;
and carrying out scaling processing on the normalized city correlation factor data so that the scaling processing result and the first price data belong to the same range.
Optionally, the first building module 420 is further configured to:
determining whether a competition impact factor is present;
if not, training and learning by utilizing the preprocessed city related factor data to obtain a first multi-factor coefficient model so as to output multi-factor coefficients;
if yes, obtaining second price data corresponding to the competition influence factors;
and training and learning by utilizing the preprocessed city correlation factor data and the second price data to obtain a second multi-factor coefficient model so as to output multi-factor coefficients.
Optionally, as shown in fig. 5, a further structural schematic diagram of the apparatus 400 for constructing the multi-factor logistics product price model is shown. Before the second building block 430, the method further includes:
and a transformation module 440, configured to perform logarithmic transformation on the first price data and the component data.
The device for constructing the multi-factor logistics product price model provided by the embodiment can execute the embodiment of the method, and the implementation principle and the technical effect are similar, and are not described herein again.
Fig. 6 is a schematic structural diagram of a computer device according to an embodiment of the present invention. As shown in fig. 6, a schematic structural diagram of a computer system 600 suitable for implementing a terminal device or a server according to an embodiment of the present application is shown.
As shown in fig. 6, the computer system 600 includes a Central Processing Unit (CPU)601 that can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM)602 or a program loaded from a storage section 508 into a Random Access Memory (RAM) 603. In the RAM 603, various programs and data necessary for the operation of the system 600 are also stored. The CPU 601, ROM 602, and RAM 603 are connected to each other via a bus 604. An input/output (I/O) interface 606 is also connected to bus 604.
The following components are connected to the I/O interface 605: an input portion 606 including a keyboard, a mouse, and the like; an output portion 607 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage section 608 including a hard disk and the like; and a communication section 609 including a network interface card such as a LAN card, a modem, or the like. The communication section 609 performs communication processing via a network such as the internet. The driver 610 is also connected to the I/O interface 605 as needed. A removable medium 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 610 as necessary, so that a computer program read out therefrom is mounted in the storage section 608 as necessary.
In particular, according to embodiments of the present application, the process described above with reference to fig. 1 may be implemented as a computer software program. For example, embodiments of the present application include a computer program product comprising a computer program embodied on a computer readable medium, the computer program containing program code for performing the above-described method of constructing a multi-factor logistics product pricing model. In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 609, and/or installed from the removable medium 611. The above-described functions defined in the system of the present application are executed when the computer program is executed by the Central Processing Unit (CPU) 601.
It should be noted that the computer readable medium shown in the present application may be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present application, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In this application, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units or modules described in the embodiments of the present application may be implemented by software or hardware. The described units or modules may also be provided in a processor. The names of these units or modules do not in some cases constitute a limitation of the unit or module itself. The described units or modules may also be provided in a processor, and may be described as: a processor includes an acquisition module, a first build module, and a second build. The names of the units or modules do not limit the units or modules, for example, the acquiring module can be further described as "first price data and component data for acquiring the city-related factor data and the at least one logistics product in the flow direction".
As another aspect, the present application also provides a computer-readable medium, which may be contained in the electronic device described in the above embodiments; or may exist separately without being assembled into the electronic device. The computer readable medium carries one or more programs which, when executed by an electronic device, cause the electronic device to implement the method of constructing a multi-factor logistics product price model as in the above embodiments.
For example, the electronic device may implement the following as shown in fig. 1: step 110, acquiring city related factor data and first price data and quantity data of at least one logistics product flowing in the direction; the city correlation factor data comprises economic indexes of at least two cities, the distance between any two cities and city grades; step 120, training and learning by using the city correlation factor data to obtain a multi-factor coefficient model so as to output multi-factor coefficients; and 130, training and learning by using the first price data, the quantity data and the multi-factor coefficient to obtain a logistics product price model. As another example, the electronic device may implement the various steps as shown in fig. 2.
It should be noted that although in the above detailed description several modules or units of the device for action execution are mentioned, such a division is not mandatory. Indeed, the features and functionality of two or more modules or units described above may be embodied in one module or unit, according to embodiments of the application. Conversely, the features and functions of one module or unit described above may be further divided into embodiments by a plurality of modules or units.
Moreover, although the steps of the methods herein are depicted in the drawings in a particular order, this does not require or imply that the steps must be performed in this particular order, or that all of the depicted steps must be performed, to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step execution, and/or one step broken down into multiple step executions, etc.
Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware.
The above description is only a preferred embodiment of the application and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention herein disclosed is not limited to the particular combination of features described above, but also encompasses other arrangements formed by any combination of the above features or their equivalents without departing from the inventive concept. For example, the above features may be replaced with (but not limited to) features having similar functions disclosed in the present application.
Claims (10)
1. A method for constructing a multi-factor logistics product price model is characterized by comprising the following steps:
acquiring city related factor data and first price data and first quantity data of at least one logistics product in a flow direction; the city correlation factor data comprises economic indexes of at least two cities, the distance between any two cities and city grades;
training and learning by utilizing the city correlation factor data to obtain a multi-factor coefficient model so as to output multi-factor coefficients;
and training and learning by using the first price data, the first quantity data and the multi-factor coefficient to obtain a logistics product price model.
2. The method for constructing the multi-factor logistics product price model according to claim 1, wherein the training and learning by using the city correlation factor data to obtain the multi-factor coefficient model comprises:
preprocessing the city correlation factor data;
and training and learning by utilizing the preprocessed city correlation factor data to obtain the multi-factor coefficient model.
3. The method for constructing a multi-factor logistics product price quantity model according to claim 2, wherein the preprocessing the city-related factor data comprises at least one of:
filling null values into the city correlation factor data;
carrying out abnormal value detection on the city correlation factor data;
and carrying out scaling processing on the city correlation factor data.
4. The method for constructing the multi-factor logistics product price quantity model according to claim 3, wherein the scaling process of the city-related factor data comprises:
carrying out normalization processing on the city correlation factor data;
and carrying out scaling processing on the normalized city correlation factor data so that the scaling processing result and the first price data belong to the same range.
5. The method for constructing the multi-factor logistics product price model according to claim 2, wherein the training and learning of the preprocessed city related factor data to obtain the multi-factor coefficient model comprises:
determining whether a competition impact factor is present;
if not, training and learning by utilizing the preprocessed city related factor data to obtain a first multi-factor coefficient model so as to output the multi-factor coefficient;
if yes, obtaining second price data corresponding to the competition influence factors;
and training and learning by utilizing the preprocessed city correlation factor data and the second price data to obtain a second multi-factor coefficient model so as to output the multi-factor coefficient.
6. The method for constructing the multi-factor logistics product price model according to any one of claims 1-5, wherein before the logistics product price model is obtained by training and learning by using the first price data, the first quantity data and the multi-factor coefficient, the method further comprises:
logarithmically transforming the first price data and the first volume data.
7. A multi-factor logistics product price model application method is characterized by comprising the following steps:
receiving first price data or a first piece of volume data;
and determining the component data corresponding to the first price data or the price data corresponding to the first component data according to a function mapping relation between the price data and the component data, wherein the function mapping relation is that the product of the price elasticity coefficient power of the price data and a multi-factor coefficient is used for representing the component data.
8. The multi-factor product price model application method of claim 7, wherein the multi-factor coefficients are determined by a pre-constructed multi-factor coefficient model.
9. The multi-factor product price model application method of claim 8, wherein the multi-factor coefficient model is constructed by:
acquiring city correlation factor data, wherein the city correlation factor data comprises economic indexes of at least two cities, a distance between any two cities and city grades;
and training and learning by utilizing the city correlation factor data to obtain the multi-factor coefficient model.
10. The method for applying the multi-factor product price model according to claim 9, wherein the training and learning by using the city correlation factor data to obtain the multi-factor coefficient model comprises:
determining whether a competition impact factor is present;
if not, training and learning by utilizing the city related factor data to obtain a first multi-factor coefficient model so as to output the multi-factor coefficient;
if yes, obtaining second price data corresponding to the competition influence factors;
and training and learning by utilizing the city correlation factor data and the second price data to obtain a second multi-factor coefficient model so as to output the multi-factor coefficient.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910446921.3A CN112001738A (en) | 2019-05-27 | 2019-05-27 | Method for constructing multi-factor logistics product price model and application method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910446921.3A CN112001738A (en) | 2019-05-27 | 2019-05-27 | Method for constructing multi-factor logistics product price model and application method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112001738A true CN112001738A (en) | 2020-11-27 |
Family
ID=73461340
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910446921.3A Pending CN112001738A (en) | 2019-05-27 | 2019-05-27 | Method for constructing multi-factor logistics product price model and application method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112001738A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112686703A (en) * | 2020-12-31 | 2021-04-20 | 长沙市到家悠享网络科技有限公司 | Automatic generation and query method for national household industry price and electronic equipment |
-
2019
- 2019-05-27 CN CN201910446921.3A patent/CN112001738A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112686703A (en) * | 2020-12-31 | 2021-04-20 | 长沙市到家悠享网络科技有限公司 | Automatic generation and query method for national household industry price and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110969285B (en) | Prediction model training method, prediction device, prediction equipment and medium | |
| CN113777552B (en) | Method and system for predicting metering error of capacitive voltage transformer | |
| CN114282670A (en) | Neural network model compression method, device and storage medium | |
| CN112561082A (en) | Method, device, equipment and storage medium for generating model | |
| CN113360711A (en) | Model training and executing method, device, equipment and medium for video understanding task | |
| CN112684396A (en) | Data preprocessing method and system for electric energy meter operation error monitoring model | |
| CN114742237A (en) | Federal learning model aggregation method and device, electronic equipment and readable storage medium | |
| CN112001738A (en) | Method for constructing multi-factor logistics product price model and application method thereof | |
| CN117593115A (en) | Feature value determining method, device, equipment and medium of credit risk assessment model | |
| CN117128950A (en) | Point cloud map construction method and device, electronic equipment and storage medium | |
| CN111986552B (en) | Map data quality freshness acquisition method, device and storage medium | |
| CN108764301B (en) | A kind of distress in concrete detection method based on reversed rarefaction representation | |
| CN116128135A (en) | Data processing method and device, electronic equipment and storage medium | |
| CN112966971B (en) | Project workload assessment method and device | |
| CN112700270B (en) | Score data processing method, device, equipment and storage medium | |
| CN114968821A (en) | Test data generation method and device based on reinforcement learning | |
| CN114625821A (en) | Perception data processing method, electronic device and program product | |
| CN114037052A (en) | Training method and device for detection model, electronic equipment and storage medium | |
| CN113807391A (en) | Task model training method and device, electronic equipment and storage medium | |
| CN108876393B (en) | Method and device for judging user risk degree | |
| CN112016944A (en) | Cost component model construction method, model application method and storage medium | |
| CN111507548B (en) | Method and system for carrying out regional difference correction on logistics service evaluation | |
| CN118096242A (en) | Prediction method, prediction device, electronic equipment and storage medium | |
| CN112001665A (en) | Logistics cost dividing method and device | |
| CN115859096A (en) | Blood glucose prediction model training method, device and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |